POVERTY AND THE EARNED INCOME TAX
CREDIT
Mindy Ault & Cherrie Bucknor
Abstract: Using the March 2012 Annual Social and Economic Supplement
to the Current Population Survey, we estimated the effect of receiving the Earned
Income Tax Credit (EITC) on the after-tax poverty status of low-wage earners.
was below the federal poverty line before taxes in order to determine whether
the EITC lifts them out of poverty. Our probit regression models showed that
those who received the credit were 53.3 percentage points more likely to be
over the poverty line after taxes compared to those who did not receive the
credit. Moreover, an additional $100 in the value of the Earned Income Tax
Credit increases the likelihood of being over the poverty line after taxes by .94
percentage points.
INTRODUCTION: THE EARNED INCOME TAX CREDIT
As the U.S. economy continues to struggle after the burst of the housing bubble in
issue. The working poor, as a group, are disproportionately affected by economic
downturns. In 2012, there were 46.5 million people in poverty.1 The U.S. Bureau of
the labor force (that is, working or looking for work) but whose incomes still fell below
1975 as a means of offsetting payroll taxes paid by poor wage earners.
The primary purpose of the EITC is to lift people out of poverty and to help those
only marginally above the poverty line to afford what they need.3 Another goal of the
EITC is to encourage labor market participation by rewarding employment. The credit
federal income tax. The EITC is the largest cash transfer welfare program in the U.S.,
1 Carmen DeNavas-Walt, Bernadette D. Proctor, and Jessica C. Smith, Income, Poverty, and Health
Insurance Coverage in the United States: 2012, P60-245
2013), 1. Accessed November 1, 2013. http://www.census.gov/prod/2013pubs/p60-245.pdf.
2 U.S. Bureau of Labor Statistics,
. (Report 1041)(Washington: U.S.
Journal of Poverty 10, no. 3 (2006).
The Public Purpose . Vol XII . 2014 [ 1 ]
Mindy Ault & Cherrie Bucknor
2003.4
from an expenditure that was 1.6 percent of the total federal budget.5
This study examined the effect of receiving the EITC on the after-tax poverty
status of low-wage earners and their families. We hypothesized that receiving the EITC
increases the likelihood that a family will be over the poverty line after taxes. We also
anticipated that the likelihood of being over the poverty line after taxes will increase as
the value of the EITC increases.
LITERATURE REVIEW
Effectiveness
The EITC is widely hailed as a successful program that helps to raise low-wage
workers and their families out of poverty and encourage work.6 Holt found that in 2003
the EITC raised 4.4 million people over the federal poverty line, over half of whom
were children. He further noted that the EITC moves more children out of poverty than
any other program, and that without it, the number of children living in poverty would
be 25 percent higher.7 Using pre-tax income data from the Current Population Survey
Annual Social and Economic Supplement to calculate poverty status before and after
the tax credit, Meyer concluded that in 2007 the EITC lowered the poverty rate overall
by 10 percent and the poverty rate for children by 16 percent.8
In theory the EITC supports all low-wage workers; however, in practice, it is
9
Meyer noted that although the
EITC is not expressly targeted to single mothers, the income ranges and the structure
of the credit—which increases with the number of children in the household—direct
it implicitly toward single mothers.10 Additionally, because the credit amount grows
larger with the number of children in the household, most EITC dollars are distributed
to families with at least two qualifying children.11
The Future of Children 17, no. 2 (2007).
collection.action?collectionCode=BUDGET&browseath=Fiscal+Year+2003&isCollapsed=false&leafLevelB
rowse=false&isDocumentResult=true&ycord=70
Metropolitan Policy Program,
The Brookings Institution (2006), 26. Accessed October 7, 2013. http://www.brookings.edu/research/
Tax Policy and the Economy, Vol. 24, ed. J.R. Brown (Chicago:
University of Chicago Press, 2010).
[ 2 ] The Public Purpose . Vol XII . 2014
Poverty and the Earned Income Tax Credit
Families below the poverty line are not the only ones who are helped by the
approximately 125 percent of the federal poverty threshold.12 By the same token, there
are many living in poverty who are not helped by the credit. Holt further found that
approximately one-third of households under the poverty threshold do not qualify for
the EITC due to age (with many being elderly) or childlessness, and about the same
number do not qualify for the credit because they lack earnings. He noted that of all
households below the poverty threshold, only about 35 percent are eligible for the
EITC.13
Disincentives to Marry and Lack of Public Awareness
One of the drawbacks frequently cited about the EITC is that it presents a
demonstrable disincentive to marry, even after marriage penalty reductions, which
were introduced after 2005).14 Hoffman observed that the primary causes of the penalty
levels.15 Berlin found that that the penalty especially affected married couples where
both spouses earned similar wages.16 This suggests that not only are couples who are
individually eligible for the credit incentivized not to marry, but also that couples who
Another issue affecting participation rates for eligible workers is that of awareness.
There are discrepancies in awareness of the credit by race and by education level. Maag
reported that in 2001, only 68 percent of low-income Black parents and a mere 27
percent of low-income Hispanic parents were aware of the EITC, and that parents with
17
Some Arguments against the EITC
The most commonly invoked argument against the EITC is the issue of
for it.18 The EITC is regulated by the IRS but receives less monitoring than other cash
transfer programs. This may explain why there are higher error rates for the EITC than
for some other programs.19 Some errors include the following: claiming children who
to provide necessary documentation, and mistakes occurring due to language barriers;
all of these which may lead to noncompliance.20 To combat these issues the IRS has
12 Ibid.
13 Ibid.
University of
Delaware
Department of Economics Working Paper Series No. 2003-01 (2003). Accessed October 7, 2013. http://
Tax Notes 106, no. 11 (2005): 1323.
20 Ibid.
The Public Purpose . Vol XII . 2014 [ 3 ]
Mindy Ault & Cherrie Bucknor
on tax forms.21
Another argument against the EITC is that it does not go far enough to address
He found that declines in wages for single mothers—caused by an increase in the labor
supply as a response to the EITC—offset increases in work hours, making the net effect
of the credit inconsequential.22 In addition, Alstott noted that use of the federal poverty
reasonable for even a very low standard of living. She cited 200 percent of the poverty
line as a more reasonable measure and found that even the maximum EITC amount of
of the poverty threshold for any size household, even for full-time year-round work.23
wage workers while they are employed; it is this class of workers that is more likely
to experience interruptions in employment due to layoffs, voluntary separation, and
disability, than workers earning higher wages.24
Arguments for and against the EITC are important in any analysis of the
families who have participated in it, and some economists and scholars argue instead
custodial parents. Berlin posited that the EITC would be a greater contributing factor
to social equity and an important incentive toward employment if it were targeted
toward individuals regardless of marital or custodial status and calculated based on
individual rather than joint income.25 Berlin also noted that low-wage noncustodial
parents—fathers, in particular—would be in a much stronger position to pay child
support to their custodial co-parents if they were earning this credit as a supplement to
their wages.26 The Center for Budget and Policy Priorities has also suggested several
improvements to the EITC. The Center believes that the EITC should be strengthened
for childless workers. This can be done by lowering the minimum age, and raising the
maximum credit and phase-in rate for these workers.27
THEORETICAL BACKGROUND AND MODEL
The federal EITC began in 1975 as a program to reduce poverty for low-income
21 Ibid.
American Economic Journal: Economic Policy 2, no. 1 (2010).
Law and Contemporary Problems 73, no. 1
(2010).
24 Ibid.
26 Ibid.
[ 4 ] The Public Purpose . Vol XII . 2014
Poverty and the Earned Income Tax Credit
poverty line and further supplement the income of the near poor to a point at which
28
This acknowledges the fact that
there are millions of people in the U.S. that are part of the working poor. Low wages
7.2 percent of workers aged 18-64 were part of the working poor in 2011.29 In order
to qualify for the EITC, an individual or married couple must have wages from work,
requirements.30
The EITC is a refundable tax credit that increases for each dollar amount that the
individual or couple earns from work. It is adjusted for the number of eligible children
family size), the individual or couple is eligible for the maximum credit. As income
increases even further, the individual or couple continues to receive the maximum
credit until they reach the beginning of the phase-out range, where a reduction in the
credit occurs for each additional dollar amount of income until it reaches zero.31
there have been a number of changes that have substantially increased the maximum
credit available and widened the range of eligible incomes.32 The last expansion
of the EITC occurred as a result of changes made in the American Recovery and
Reinvestment Act of 2009 (ARRA). ARRA created a category for families with three or
more children and broadened the phase-in range for married couples.33 These changes
were intended to be a temporary response to the Great Recession. The American
Taxpayer Relief Act of 2012 (ATRA) extended the ARRA changes through 2017.34
As mentioned previously, the EITC has been shown to lift families out of poverty,
with Holt showing that in 2003, 4.4 million people were thus assisted by the credit,
and Meyer observing, in CPS income data for 2007, that the overall poverty rate was
reduced by at least 10 percent.35 The EITC acts as a supplement to the low wages
that many families receive. In theory, providing a supplement to workers’ incomes
will increase their after-tax income. This increases their income-to-poverty ratio and
increases the likelihood that they will be above the poverty threshold after taxes.
A theoretical linear model of the effect of the EITC on poverty status would look
as follows:
(1) Pr(yi=1|xi
x
0
1 i
i
i
29 DeNavas-Walt, Proctor, and Smith, Income, Poverty, and Health, http://www.census.gov/prod/2013pubs/
p60-245.pdf.
http://www.irs.gov/pub/irs-pdf/p3211.pdf.
taxfacts/displayafact.cfm?Docid=36.
American Economic Review 102, no. 5 (2012); Y. Lim and C. Lemieux,
Journal of Sociology & Social Welfare
33 American Recovery and Reinvestment Act of 2009. P.L. 111-5.
34 American Taxpayer Relief Act of 2012. P.L. 112-240.
The Public Purpose . Vol XII . 2014 [ 5 ]
Mindy Ault & Cherrie Bucknor
Where y = a dummy variable for having an income above the poverty line after
taxes, x = a dummy variable for EITC receipt, is a vector of other covariates, and is
an error term. This theoretical model is useful because it measures the probability that
an individual or household would be out of poverty after taxes. In our study, we created
two models—one with a continuous measure of EITC value and another with a binary
measure of EITC receipt. Our hypothesis was that receiving the EITC increases the
likelihood that a family will be over the poverty line after taxes, and that the likelihood
of being over the poverty line after taxes increases as the value of the EITC increases.
Data
The data for this study come from the Current Population Survey (CPS),
Annual Social and Economic Supplement (ASEC) of 2012. The U.S. Census Bureau
ASEC is the March Supplement to the CPS. It is a nationally representative sample
and members of the Armed Forces living in civilian housing units on a military base
36
Data is collected through in-person and
telephone interviews. While the basic CPS monthly survey mostly includes data
on employment, the ASEC provides more detailed data on income, demographics,
program participation, taxes, and work experience. The income and tax information
are asked for the previous year. This dataset is well suited for this study because it is
the most recent data available and includes variables that measure total family income,
EITC receipt, tax liability before and after credits, and poverty thresholds. It also
includes data on respondents’ race, ethnicity, education, marital status, and age.
The full sample contains 201,398 individuals in 74,383 unique households. To
obtain the analytic sample, we only included observations for the reference person in
each household. The reference person in the survey is the most knowledgeable adult or
head of household. The data from the reference person contains all of the information
relevant to this study—most notably, family income, EITC receipt, and tax liability. Of
the individuals in the full sample, 74,383 were heads of household. The sample was
further restricted to heads of household whose family was under the poverty threshold
before taxes (10,598). Of these heads of household, only 4,874 had earnings from
4,874 heads of household.
The dependent variable is a poverty dummy variable where 1 = above the poverty
line after taxes, and 0 = below the poverty line after taxes. To determine poverty status
variables from the dataset. They were: total family income before taxes, federal
income tax liability after credits were applied, state income tax liability after credits
were applied, and poverty threshold for each family (given family size); these are all
36 U.S. Census Bureau, Current Population Survey, 2012 Annual Social and Economic Supplement
Technical Documentation
2013. http://www.census.gov/prod/techdoc/cps/cpsmar12.pdf
[ 6 ] The Public Purpose . Vol XII . 2014
Poverty and the Earned Income Tax Credit
continuous variables measured in dollars. The formula to determine after-tax liability
was as follows:
(2) after tax income = total family income – federal income tax liability after
credits –
state income liability after credits
We used the after-tax income variable along with poverty threshold to create a
new variable that measured each family’s poverty ratio after taxes. The formula was as
follows:
(3) poverty ratio after tax = after tax income / poverty threshold
This was a continuous variable. We recoded this into a dummy variable as stated
above.
The independent variable of interest was a continuous EITC variable representing
the dollar amount of EITC the family received. We also created a dummy variable
where 1 = received the EITC, and 0 = did not receive the EITC.
The control variables included in the model were: a continuous variable for age,
dummy variables measuring education (less than high school, high school, and postsecondary education); ethnicity (Hispanic); race (white, black, Asian / Hawaiian /
Appendix 1 displays the descriptive statistics for the variables included in our
analysis. There are 4,874 heads of household in the sample. 3,306 are female, and
1,838 are male. The average age is 38, with the lowest being 15 and the highest being
85. There were 1,350 people who did not complete high school, 1,634 who earned a
high school diploma, and 1,890 who had some sort of post-secondary education (some
college, associate’s degree, bachelor’s degree, master’s degree). The average amount
heads of households who were above the poverty line after taxes, and 3,582 who
remained under the poverty line after taxes. There were also 1,730 heads of households
who were married, and 3,144 who were either single, divorced, or widowed. 3,434
heads of household reported their race as white; 945 as black; 247 as Asian, Hawaiian,
Empirical Method
Our dependent variable is a binary variable that describes poverty status after
taxes. Accordingly, we estimated probit models to determine the effect of receiving
the EITC on poverty status after taxes. To test the robustness of our model, we also
estimated a linear probability model (LPM). Our preferred model is a probit regression
because our dependent variable had only two values, coded as 0 and 1. The LPM is
usually not ideal because it can result in predicted probabilities outside of the [0,1]
range. The LPM also keeps the average partial effects constant and has an inherent
problem of heteroskedasticity. While the problem of heteroskedasticity can be resolved
The Public Purpose . Vol XII . 2014 [ 7 ]
Mindy Ault & Cherrie Bucknor
or above 1, the LPM will not produce a good approximation of the true data generating
process (DGP).37
Probit models estimated by maximum likelihood ensure that the conditional
probability is bounded by 0 and 1. We report the average partial effect of a continuous
x (in this case the continuous or binary measure of EITC) as:
−1
[
=1
1
(
0
+
1
+
2
)]
Where is the standard normal probability distribution function (PDF), x is the
continuous or binary EITC variable, and is an index of covariates measuring age,
race, ethnicity, education, gender, and marital status. The average partial effects are
directly comparable to OLS estimates.38
RESULTS
Models with a Continuous EITC Variable
The results of our probit regression model of the effect of the continuous EITC
variable on after-tax poverty status are shown in Appendix 2. The results indicate that
poverty status (p<.001). The average partial effect of the EITC variable was .000094.
percentage points more likely to be above the poverty line, holding all else constant. In
addition, an increase in EITC equal to one standard deviation is predicted to make the
head of household 19.93 percentage points more likely to be above the poverty line,
holding all else constant. It is important to note, however, that this result assumes a
fairly inelastic labor demand, which does not allow for wages declining in response to
the overall increase in income for wage earners from the EITC.
Heads of household who earned a high school diploma were 4.35 percentage
points more likely to be above the poverty line after taxes compared to those who did
not complete high school. Attending at least some college made heads of household
5.07 percentage points more likely than high school graduates to be above the poverty
line. In addition, Hispanic heads of household were 3.73 percentage points more likely
than non-Hispanic heads of household to be above the poverty line. Our probit model
had an of .3751, so the model explains 37.51 percent of the variation in after-tax
We also ran a linear probability model, which is easier to interpret than probit
37 Jeffrey M. Wooldridge, Introductory Econometrics: A Modern Approach, 5th Edition (Mason, OH: SouthWestern Cengage Learning, 2013), 251.
38 See Wooldridge p. 583-587 for more on probit models.
[ 8 ] The Public Purpose . Vol XII . 2014
Poverty and the Earned Income Tax Credit
in EITC credit makes the head of household 1.3 percentage points more likely to be
above the poverty line after taxes, which is very similar to the probit average partial
effects.
(p=.001). Having a high school diploma makes a head of household 4.8 percentage
points more likely to be above the poverty line after taxes compared to those who
.05636; having at least some post-secondary education makes a head of household
5.6 percentage points more likely to be above the poverty line compared to those who
value of -.03933. Hispanic heads of household were 3.9 percentage points less likely
than non-Hispanics to be above the poverty line after taxes. Race, female, married, and
Our linear probability model had an of .3844, indicating that the model
explains 38.44 percent of the variation in after-tax poverty status. 81.99 percent of the
model.
Models with a Binary EITC Variable
We ran a probit model using a binary independent variable (EITC receipt
versus non-receipt) in place of the continuous variable (see Appendix 3). In this
model, the pseudo was 0.1484, indicating that 14.84 percent of the variation in aftertax poverty status could be explained by the independent variables in the model. This
model resulted in average partial effects (APEs) of EITC receipt, female gender,
average partial effect of EITC receipt was .533 (p<.001), indicating that, according to
this model, receipt of the EITC increases the likelihood of being out of poverty by 53.3
percentage points compared to not receiving the credit, holding all else constant.
inelastic labor demand, earning a high school diploma increases the likelihood of being
out of poverty by 5.2 percentage points; that completing at least some post-secondary
education increases this likelihood by 3.8 percentage points; and that Hispanic ethnicity
increases the likelihood by 4.8 percentage points. Further, being female and being
married both increase the likelihood of being out of poverty after taxes by 5.6 and
14.7 percentage points, respectively. As age increases, the chances of being out of
poverty after taxes are shown to decrease by .48 percentage points. 73.59 percent of the
predicted probabilities were correctly predicted.
In a linear probability model using the binary EITC dependent variable, the
was 0.1213, indicating that 12.13 percent of the variation in after-tax poverty status
EITC receipt (.261), female gender (.061), married status (.162), and age (-.004) were
school diploma (.055), post-secondary education (.037), and Hispanic ethnicity (.052)
The Public Purpose . Vol XII . 2014 [ 9 ]
Mindy Ault & Cherrie Bucknor
credit increases the likelihood of being out of poverty after taxes by 26.1 percentage
points and were half the size of those obtained from the probit model. 73.49 percent of
model.
DISCUSSION
the after-tax poverty status of low-wage workers, and that the amount of EITC received
is related the after-tax poverty status. Therefore, both of our hypotheses were supported
of the continuous EITC variable was very small (.000094). In this model, there is no
people in our sample. An additional dollar in EITC makes a head of household .0094
percentage points more likely to be above the poverty line after taxes. Even when we
of household .94 percentage points more likely to be above the poverty line. Based on
points more likely to be above the poverty line after taxes. While these results suggest
that increasing the value of the EITC will lead to an increased likelihood of being
above the poverty level, it is not realistic to recommend that the EITC be increased by
A probit model measuring the effect of a binary variable of EITC receipt
predicted an average partial effect of .533. According to this model, receiving the EITC
increases the likelihood of being above the poverty line after taxes by 53.33 percentage
points compared to those who did not receive the EITC, all else being equal. Since all
of the heads of household in the sample were in poverty before taxes, receiving the
EITC is associated with an increased likelihood of being lifted out of poverty after
taxes. These results support our hypothesis that receiving the EITC will increase the
likelihood of being above the poverty line after taxes. However, our OLS model found
that the partial effect of EITC receipt on after-tax poverty status was half that of our
probit average partial effect. This large difference is unusual, given the fact that LPM
and probit average partial effects are usually very similar.
Our analysis suffered from several limitations. One limitation is that we
assumed that people who received the EITC and people who did not receive the EITC
were equivalent in all other respects after controlling for the covariates in our model.
There are reasons why this may not be correct. Some of the people who did not receive
found that of all households below the poverty threshold, only about 35 percent are
eligible for the EITC.39 Eligibility requirements for the EITC include having a valid
or being married to a U.S. citizen or resident alien for the entire tax year, and having
[ 10 ] The Public Purpose . Vol XII . 2014
Poverty and the Earned Income Tax Credit
a qualifying child.40 If an individual does not have a qualifying child, they must be
year.41 Due to the limitations of our dataset, we were unable to restrict our analytic
sample to only those who were eligible for the EITC.
Another limitation is that we were unable to control for the receipt of a state
EITC. In 2012, 24 states and the District of Columbia provided a supplemental EITC
as a percentage of the federal EITC that a household is eligible for.42 This can have
an effect on the after-tax poverty status of households. Our dataset did not include
information on state EITC receipt, and we were therefore unable to control for this in
our models.
labor demand that does not result in reduced wages in response to the EITC. We opted
labor supply increases in response to the EITC take place among single mothers.43 For
the parameters of this study, we included all household types, not just single-mother
families.
A suggestion for further study would be to test the effects of the EITC on
other outcomes and to use a similar model but with an assumption of wages adjusted
for a more elastic labor demand. Researchers could estimate models that predict
whether workers respond to the work incentives of the EITC by working additional
hours. Another possibility would be to determine if receiving the EITC has an effect
on child health or school performance. Propensity score matching based on EITC
receipt may also be used. In addition, although the EITC has a primary goal of lifting
individuals and families out of poverty, research has shown that it has been used
poverty threshold, further research should be done to determine how receiving a wage
supplement in the form of the EITC can affect the well-being of families who are low
including additional control variables such as other tax credits claimed, unemployment
rate, and cash assistance received.
41 Ibid.
December 29, 2013, http://www.cbpp.org/cms/?fa=view&id=2506.
The
Quarterly Journal of Economics
The American Economic Review 92, no. 2
(2002).
The Public Purpose . Vol XII . 2014 [ 11 ]
Mindy Ault & Cherrie Bucknor
APPENDIX 1: DESCRIPTIVE STATISTICS
Variables
Mean
(SD)
Not in poverty after taxes
.242
(.428)
Value of EITC
2,051.37
(2,120.43)
EITC (1=Received)
.8010
(.3993)
Education
High School
.3352
(.4721)
Post-Secondary Education
.3991
(.4898)
Hispanic
.2675
(.4427)
.0475
(.2127)
Black
.2212
(.4151)
Native American/Alaska Native
.0190
(.1364)
Mixed Race
.0220
(.1467)
Female
.6156
(.4865)
Married
.3299
(.4702)
Age
37.93
(13.21)
N
4,874
Note: Sampling weight was used
[ 12 ] The Public Purpose . Vol XII . 2014
Poverty and the Earned Income Tax Credit
APPENDIX 2 – REGRESSION RESULTS WITH CONTINUOUS EITC VARIABLE
Variables
Probit
Probit APE
OLS
EITC value
0.0005
(0.0000)***
0.0001
(0.0000)***
0.0001
(0.0000)***
High School
0.2289
(0.0760)***
0.0435
(0.0143)***
0.0488
(0.0152)***
Post-Secondary
Education
0.2673
(0.0777)***
0.0507
(0.0146)***
0.0564
(0.0146)***
Islander
-0.0222
(0.1392)
-0.0042
(0.0264)
-0.0030
(0.0282)
Black
-0.1097
(0.0730)
-0.0208
(0.0138)
-0.0232
(0.0137)*
Native American
Alaska Native
0.0646
(0.2392)
0.0123
(0.0455)
-0.0024
(0.0504)
Mixed Race
-0.2573
-0.2145
-0.0489
(0.0407)
-0.0455
(0.0358)
Hispanic
-0.1964
(0.0704)***
-0.0373
(0.0134)***
-0.0393
(0.0145)***
Female
0.0395
(0.0595)
0.0075
(0.0113)
-0.0010
(0.0108)
Married
-0.0006
(0.0689)
-0.0001
(0.0131)
-0.0170
(0.0147)
Age
-0.0032
(0.0025)
-0.0006
(0.0005)
-0.0000
(0.0003)
Constant
-2.0879
(0.1306)***
-0.0375
(0.0206)*
0.384
Pseudo
0.375
N
4,874
4,874
4,874
Notes: *** p<0.01, ** p<0.05, * p<0.1, robust standard errors in parentheses, a sampling
weight was used.
The Public Purpose . Vol XII . 2014 [ 13 ]
Mindy Ault & Cherrie Bucknor
APPENDIX 3: REGRESSION RESULTS WITH DUMMY EITC VARIABLE
Variables
Probit
Probit APE
OLS
EITC
2.0020
(0.1771)***
0.5331
(0.0464)***
0.2607
(0.0097)***
High School
0.1938
(0.0653)***
0.0516
(0.0173)***
0.0547
(0.0179)***
Post-Secondary
Education
0.1419
(0.0662)**
0.0378
(0.0176)**
0.0367
(0.0174)**
Asian/Hawaiian/
0.0317
(0.1153)
0.0084
(0.0307)
0.0103
(0.0309)
Black
0.0777
(0.0654)
0.0207
(0.0174)
0.0186
(0.0174)
Native American
Alaska Native
0.0885
(0.2255)
0.0236
(0.0601)
0.0202
(0.0672)
Mixed Race
-0.1677
(0.1762)
-0.0447
(0.0469)
-0.0480
(0.0414)
Hispanic
0.1817
(0.0597)***
0.0484
(0.0158)***
0.0516
(0.0171)***
Female
0.2115
(0.0524)***
0.0563
(0.0139)***
0.0608
(0.0132)***
Married
0.5505
(0.0535)***
0.1466
(0.0137)***
0.1618
(0.0158)***
Age
-0.0181
(0.0022)***
-0.0048
(0.0006)***
-0.0040
(0.0004)***
Constant
-2.3898
(0.1862)***
0.0417
(0.0234)*
0.121
Pseudo
0.148
N
4,874
4,874
4,874
Notes: *** p<0.01, ** p<0.05, * p<0.1, robust standard errors in parentheses, a sampling
weight was used.
[ 14 ] The Public Purpose . Vol XII . 2014
Poverty and the Earned Income Tax Credit
BIBLIOGRAPHY
Law and Contemporary
Problems 73, no. 1 (2010): 285-313.
American Recovery and Reinvestment Act of 2009. P.L. 111-5.
American Taxpayer Relief Act of 2012. P.L. 112-240.
The Future of Children 17, no.
2 (2007): 17-42.
cms/?fa=view&id=2506.
www.cbpp.org/cms/?fa=view&id=3991.
American Economic Review
102, no. 5 (2012): 927-1956.
DeNavas-Walt, Carmen, Bernadette D. Proctor, and Jessica C. Smith. Income,
Poverty, and Health Insurance Coverage in the United States: 2012, P60-245
2013. http://www.census.gov/prod/2013pubs/p60-245.pdf.
The Quarterly Journal of Economics 111, no. 2 (1996): 605-637.
University of
Delaware Department of Economics Working Paper Series No. 2003-01 (2003):
935-941. Accessed October 7, 2013. http://graduate.lerner.udel.edu/sites/default/
Metropolitan
Policy Program, The Brookings Institution (2006). Accessed October 7, 2013.
http://www.brookings.edu/research/reports/2006/02/childrenfamilies-holt
2013. http://www.irs.gov/pub/irs-pdf/p3211.pdf.
Journal of
Sociology & Social Welfare 35, no. 2 (2008): 49-65.
The Public Purpose . Vol XII . 2014 [ 15 ]
Mindy Ault & Cherrie Bucknor
1323.
Journal of Poverty 10, no. 3 (2006): 51-68.
The American Economic Review 92, no. 2 (2002):
373-379.
Tax Policy and the Economy, Vol. 24, edited by J.R. Brown, 153-180. Chicago:
University of Chicago Press, 2010.
American Economic Journal: Economic Policy 2, no. 1 (2010):
177-208.
taxpolicycenter.org/taxfacts/displayafact.cfm?Docid=36.
U.S. Bureau of Labor Statistics.
, 2011. (Report 1041).
2013. http://www.bls.gov/cps/cpswp2011.pdf
U.S. Census Bureau. Current Population Survey, 2012 Annual Social and Economic
Supplement Technical Documentation. Washington: U.S. Government
prod/techdoc/cps/cpsmar12.pdf
from https://www.census.gov/popest/data/state/totals/2003/tables/NSTEST2003-02.pdf
Accessed February 28, 2014 from: http://www.gpo.gov/fdsys/browse/
collection.action?collectionCode=BUDGET&browseath=Fiscal+Year+2003
&isCollapsed=false&leafLevelBrowse=false&isDocumentResult=true&yco
rd=70
Wooldridge, Jeffrey M. Introductory Econometrics: A Modern Approach, 5th Edition.
Mason, OH: South-Western Cengage Learning, 2013.
[ 16 ] The Public Purpose . Vol XII . 2014